One of the most important things to know when practising law in any country is what the recognised sources of law are. The reason is obvious: if you base your case on a principle which, however sensible it appears, is not recognised as part of the law, or if you ignore anything relevant which is law, you will look a fool in court, and your client won’t be very happy with you.

In some countries, the entirety of the law is (or is supposed to be) written down in one place, where it is easy for even a non-lawyer to find. Cf the French and German civil codes. Scots law is, most emphatically, not like this at all. Neither, for that matter, are English law or Northern Irish law. It seems that comprehensive codification is simply not the way we do things in the UK. Below are listed the many and varied sources from which Scots law is drawn. Afterwards I will say a brief word on juridical typology, because Scots law is in a slightly odd position when it comes to classifying the world’s legal systems, but in order to understand why, you first need to know about the sources of law.

Common law

For continental Europeans, the most striking characteristic of the UK’s legal systems is their heavy reliance on previous court decisions – the French even call it "la superstition du cas". In some areas of law there is almost nothing to be found in the statute books, except some very short statutes dealing with minor points. Instead of reading statutes to find the law, you need to read cases: judgements given in previous disputes. In other areas, it seems as though half the subject is covered by the relevant statutes, while they are silent on the other half. Again, the law is found in the cases – but you have to be careful that the case you want to rely on has not been rendered obsolete by a more recent statute. Even in areas where statutes seem to cover everything, you ignore the case law at your peril – hidden ambiguities of language and interpretation may lurk, and it is case law that resolves them.

The theory (or, perhaps, the fiction) of common law jurisprudence goes something like this: the Common Law of Scotland has existed since time immemorial, and never changes. It is not, despite appearances, created by judges; rather, judges tell us what it is (and always has been). When one judgement overrules a previous one, the law is not being changed; it is being clarified, or “developed”. The overruling judge is in effect saying that the earlier judgement was wrong, and every subsequent decision which has relied on it was also wrong. (In practice, of course, this is very like retroactive lawmaking, and very little can be done to keep from being caught out by it.) Statutes may suppress the common law, but if those statutes are repealed, the common law revives.

Because judges build on each other’s work in developing the common law, and because the first judge to state a rule is not (supposedly) creating it ex nihilo, it is not always necessary to trace a rule to its earliest origins. Instead, when looking for a case to cite as authority for the interpretation of the law they wish to put forward, lawyers will typically choose the one with the highest authority, or the clearest example, rather than the first case that articulated the rule. For this reason, judges in the superior courts often provide in their judgements a helpful restatement or summary of the development of the law thus far, to save future lawyers and judges the hassle of digging through what may be a tangled mess of older cases.

Central to the idea of case law is the idea that like cases should be decided alike. In legal jargon this is called the principle of stare decisis (“standing by what has been decided”) or the “doctrine of precedent”. If the question before the court has been decided before, the answer is there. But of course there must be an amending mechanism, otherwise early mistakes would only ever become more entrenched and could never be fixed. Therefore, all precedent-based legal systems have a hierarchical court structure in which higher courts may overrule the earlier decisions of lower courts, and create “binding precedents” which lower courts must follow. The basic rule is that the decisions of one court bind another court if there is an appeal from the second to the first (though this breaks down once you get to the top court). There are two hierarchies in Scotland, one for criminal cases and one for civil cases (“civil” in this context means all cases other than criminal cases). I will put an overview of the Scottish court structure in the comments, as to include it here would make this post too long.


That like cases should be decided alike (unless, of course, the law has changed in the meantime) is an obvious requirement of fairness. But this raises the question of just how alike two cases have to be if the first is to determine the outcome of the second. As no two cases are identical, the judge must decide whether the differences between them, be they great or small, are relevant to the outcome. There are three basic options for a judge when a previous case appears to have a bearing on the present case:

1. He can follow the earlier case and order the same outcome.


2. He can distinguish the earlier case. This means saying that, while the earlier case was correctly decided, the facts of the two cases are dissimilar enough that the outcome should be different. (When the earlier decision is from a higher court, this is the only way to avoid being bound by it.)


3. If he has the authority, he can overrule the earlier case. This means saying that it was wrongly decided and has never in fact been good law. The effect is that the overruled case is no longer a valid precedent.

A variation on the last option occurs when the judge does not have the authority to overrule the earlier case, but is not bound by it either, such as when it came from another court at the same level in the hierarchy. He can then disapprove the earlier case and decline to follow it. The resulting contradiction will have to be dealt with by later judges, and will not be settled until the issue is raised in a higher court, or legislation intervenes.

A key part of this process is distinguishing the parts of the earlier judgement which form the rule (the ratio decidendi or simply ratio – “reason for deciding”) from other things the judge may have said which, while maybe very insightful, do not form part of the reason for the judgement (obiter dicta – “things said by the way”). Rationes may be interpreted broadly or narrowly, and the original case may give no clues as to whether the rule it expounds is supposed to apply to the current situation. Obiter dicta are sometimes approved by later judges, and indeed they may be the best available statement of the law on a given topic, but they are never formally binding in stare decisis. Lawyers sometimes argue in court about whether a given passage is obiter or not, since this determines whether or not the judge must follow it. All this is part of a day’s work for a judge in a common law system.

Cases from other countries, such as England, have persuasive value only and are not formally binding. This applies even to decisions of the House of Lords or Supreme Court, though the persuasive value of these may be very great.

Legislation

Legislation is the kind of lawmaking we are most familiar with. Indeed, pieces of legislation are usually called “laws” (though this is not strictly accurate). Here, written amendments to the existing law are approved by a recognised lawmaking body (known as a legislature) and published. Once in force, they are supposed to be “judicially noticed”, i.e. judges are supposed to treat them as law and make their decisions accordingly.

Scots law has more sources of legislation than most legal systems. Below are a few words on each of them.


This is the most important and powerful legislature in the United Kingdom. It is officially called “the Parliament of the United Kingdom”, but is often referred to as “Westminster” after the part of London in which it sits. It consists of three parts. The House of Commons is made up of representatives elected from the UK’s 650 parliamentary constituencies under the first-past-the-post system. Nearly all members of the government sit in this House, and most bills (proposed legislation) begin there.

The House of Lords is principally a revising chamber, looking at bills in more detail than the Commons has time for. It is not elected. For centuries it consisted entirely of hereditary nobility (earls, dukes and so on) and Church of England bishops, but now most of them are appointed for life by the Queen, acting on the advice of the Prime Minister (technically; in practice this means the Prime Minister appoints them). A little more democratic, but not much. However, 26 bishops and 92 noblemen still remain, out of a current total of 789 Lords. Until 1911 any bill had to be agreed by both Houses, but the House of Lords can now only delay a bill that arises in the House of Commons. A determined House of Commons can therefore defy the House of Lords. You may not be surprised to learn that the most recent example of this was the fox-hunting ban, and the one before that was to equalise the age of consent for gay sex with that for straight sex.

The third component of the UK Parliament is the Sovereign, who summons and dismisses the two Houses. As a result the three components together are sometimes called “the Queen-in-Parliament”. No bill passed by the two Houses becomes law until it receives Royal Assent. When this is granted, the bill becomes an Act of Parliament, and has the force of law. In theory this means the Queen has a veto over all legislation, as she could refuse to sign the bill. However, the last time this happened was in 1708, and it is unlikely to happen again any time soon. (But see below for Royal Assent of bills in the Scottish Parliament.)

The central thesis of UK constitutional theory is the doctrine of Parliamentary Sovereignty (or Parliamentary Supremacy). In summary, this says “What the Queen-in-Parliament enacts is law”. Unlike in many countries with a codified constitution, there is no special body of “higher law” which constrains what the legislature can do, needing more than a simple majority of 50%+1 to be amended. There are of course statutes which deal with constitutional matters, but all of them can be repealed or changed in the same way as any other statute. Because Parliament can legally do anything, the only thing it cannot do is constrain a future Parliament, since the latter too is sovereign. (Insert riddle about God making a rock so heavy he cannot lift it here.) This also means that judges cannot declare an Act of Parliament to be illegal, as US judges can do with Acts of Congress. If it’s an Act of Parliament, then by definition it’s the law.

(EU law has complicated the position of Parliamentary Sovereignty somewhat, but discussion of this is beyond the scope of this article. Suffice it to say that while EU law may sometimes beat UK law in court, the UK could always leave the EU if it wanted to.)

If an Act of Parliament is repealed (which can only be done by a later Act), it ceases to be the law. If the repealing Act is itself repealed, the first Act stays repealed – it does not revive. In this way statute law is unlike the common law, which can revive if an Act suppressing it is repealed.


In Part 1 I discussed some reasons why it was not always a good thing for Scots law that the only body that could change it was the UK Parliament, where Scots are a minority and struggle to persuade the others to spend time on law reform for Scotland. To remedy this, a devolved Scottish Parliament was created by the Scotland Act 1998, with power to make laws for Scotland in certain areas. The Scotland Act functions as a sort of miniature codified constitution for Scotland, and many of its provisions are simply copied from the unwritten rules that govern the operation of the UK Parliament.

The Scottish Parliament meets in an ugly building opposite Holyrood Palace in Edinburgh, and is therefore often referred to simply as “Holyrood”. It is unicameral (so no revising chamber like the House of Lords) and is elected under a mixed-member system, with some first-past-the-post constituency seats and some proportional representation seats for multi-member regions. Its legislation is called Acts of the Scottish Parliament.

Broadly speaking, areas of law where Scotland already had different rules are now the responsibility of the Scottish Parliament, while certain topics of national importance (such as the constitution, foreign relations, defence, taxation and nuclear power) are “reserved” to the UK Parliament. The Scotland Act contains a list of “reserved matters”, and anything that is not reserved is automatically devolved. The Scottish Parliament can repeal Acts of the UK Parliament (as indeed it must if it is to amend areas of law last updated before 1999), provided it does not transgress on reserved matters. (If you're interested to know what the Scottish Parliament can and can't do, the list of reserved matters is in Schedule 5 to the Scotland Act, and the list of statutes it can't change is in Schedule 4.)

One important limitation on the Scottish Parliament’s powers is that it may not do anything that contravenes either EU law or the European Convention on Human Rights. The creation of a legislature with limited powers means that, for the first time in the UK, judges can be put in the position of having to decide whether or not statutes are in fact the law. The Americans have been doing this since 1803, but it’s something of a novelty here.

Since the UK Parliament is sovereign and can pass any law it chooses, there is technically nothing to stop it legislating on devolved matters in Scotland. However, this would be bad manners, so by convention it asks the Scottish Parliament’s permission before doing so. The Scottish Parliament may then pass a “legislative consent motion” (or “Sewel motion”, after Lord Sewel who raised the issue in the UK Parliament when the Scotland Bill was going through the Lords) saying it doesn’t mind. This happens more often than you might think.

However, Westminster did not allow Holyrood to be completely free of the threat of a veto, even in relation to devolved matters. To do this, it rather quaintly exploited the requirement of Royal Assent (a pure formality in Westminster) for a bill to become law. Holyrood bills may not be presented for Royal Assent until four weeks after they are passed. In that time, two things can happen to prevent it. One is that one of the three “Law Officers” (top government lawyers) could refer it to the Court of Session to decide whether it is within the Parliament’s powers or not. If it is, fine, and off it goes to the Palace. If not, it cannot be presented for Royal Assent and therefore cannot become law. The other thing is that the UK Government can veto it. This can be done even if the bill is within the Parliament’s powers. The government is only supposed to do this when it thinks the bill will “adversely affect” reserved matters, but if they did it for a different reason, it would be very difficult to stop them. If a change in the law is very urgent, the four-week waiting period can be dispensed with if all the people with a power of reference or veto say they do not intend to use it.


This was Scotland’s legislature before the Union, when it was an independent country. Like the UK Parliament of today, it was sovereign and had unconstrained legislative power. It also operated in a very similar way.

Quite a few acts of the Parliament of Scotland are still in force today. The oldest dates from 1424, and goes like this:


So if you find a gold mine in your garden, you’re not in as much luck as you might have thought. The second-oldest Act is of more practical importance:


In plain English, this means that if you are tenant and the landlord changes, the new landlord can’t throw you out or raise the rent before the end of your lease. This is an extremely useful rule, despite its age.

However, some acts of the old Parliament are no longer in force despite never having been formally repealed. This is due to the “doctrine of desuetude”, which is a fancy word for “disuse”. Basically, if the rule has been ignored in practice for a long time, it falls into desuetude and ceases to be the law. Mere age is not enough, as the two Acts quoted above show – there must be longstanding contrary practice. An example is McAra v Edinburgh City Council (1913), in which the Council used powers given in an Act of 1606 to ban meetings in the street. It was held that the Act was in desuetude because people had been openly meeting in public for ages.

Desuetude probably does not apply to Acts of the UK Parliament. But it’s hard to be sure, because the doctrine doesn’t exist in English law at all, meaning less chance of a case. To find out, we would need a Scottish case in which a post-1707 Act was challenged (ideally one that applies only to Scotland). This has not yet happened.


I don’t propose to spend much time talking about the EU, because that’s a whole new topic, but basically, all EU Member States incorporate EU law into their domestic law when they join. This means that EU law is also Scots law, and can be enforced in any Scottish court. As mentioned above, the Scottish Parliament cannot make laws which contravene EU law (and it’s possible we may see the first pre-Royal Assent challenge to an Act of the Scottish Parliament with the proposed minimum pricing of alcohol, which many think is against EU law). However, the EU cannot just make any law: it is restricted to the areas of competence granted to it by the Member States in the treaties, which are mostly economic in nature.

One kind of law that emanates from the EU is the Directive. This is when the EU says that the Member States must legislate to achieve a certain goal (such as equal pay for men and women) but leaves the implementation in all its detail to the Member States themselves. This is supposed to avoid the kinds of problems that arise when laws written with one legal tradition in mind are imposed on another tradition. When the UK joined the EU, Parliament gave government ministers the power to implement directives by means of subordinate legislation (see below) to save time in Parliament. The Scotland Act gave the same power to the newly created Scottish Ministers, so they now implement directives into Scots law.


This is legislation made not by Parliament but by the government usually in the form of “statutory instruments”. This only occurs when Parliament grants power to do so. It is mostly used when minor details are likely to need to be added or amended in future. It would be a waste of time if Parliament had to debate the matter every time the government wanted to increase speeding fines to keep up with inflation, so for things like this the relevant Act of Parliament usually say e.g. “£50, or such amount as the Secretary of State may provide by order”. Statutory instruments made by the Scottish Ministers are known as “Scottish Statutory Instruments”.

Unlike Acts of Parliament, subordinate (or “secondary”) legislation can be declared illegal by judges: either for contradicting or going beyond the terms of the Act that confers the power, or for common law grounds such as “irrationality”.

Institutional writers

This is a rather strange kind of source of law. As far as I know very few if any other legal systems treat this kind of writing as authoritative. Basically these are academic textbooks dating from between the late 17th and early 19th centuries. The name is because a popular title for such books was “Institutions of the Law of Scotland” vel sim, and this owes something to the Institutes of Gaius and Justinian. It is probable that the class of Institutional Writers is now closed, so if you had hopes of becoming one yourself, you’re out of luck.

Legislation and case law will, of course, always beat a student textbook. Academics are not lawmakers. However, because in centuries past all the judges and advocates had studied using these textbooks, they were often cited in court, and their summaries of the law were incorporated into judgements and thus given authority under the doctrine of precedent. And if no precedent existed to deal with a particular problem, but one of the Institutional Writers had thought it through, their reasoning might well be adopted and thus become a precedent.

Additionally, some old cases are reported nowhere other than by these writers, so some of their writing does in fact have conventional authority. Essentially this is them acting as court reporters, since there were no comprehensive reports in those days.

Roman law

Some areas of Scots law (particularly the law of property) have their basis in Roman law. How this came about was discussed in Part 1 – there was also indirect Roman influence via the Canon Law used by church courts in the Catholic era before the Reformation. You should not by any means imagine that it is possible to pick any passage from the Digest, quote it in court, and expect the judge to immediately agree that it must represent the current law of Scotland. It is usually only when a principle of Roman law has been incorporated into Scots law by judgements of centuries past that it has any relevance.

However, very occasionally cases do arise where there is no domestic precedent and no legislation to rely on, and so it is concluded that the Roman law is indeed the current law. In Sloans Dairies Ltd v Glasgow Corporation (1977), a contract was made for the sale of a building, but before ownership was transferred, the building was damaged (but not totally destroyed) by fire. The question was whether the buyers could be held to the contract and made to pay full price for it: in the period between contract and conveyance, who bore the risk? Amazingly, the issue had never arisen in a Scottish court before, so the court relied on the rule stated in the Institutes of Justinian, which said that the buyer bore the risk and had to pay the contract price despite the damage. (This can of course be varied by agreement, which it now always is, and it doesn’t apply if the damage is in any way the seller’s fault.)

You might think that the reason why Scots law can be Roman law is because the Romans once ruled Scotland, and a few laws have come down through the millennia and never been changed since that time. This is not so: the Roman Empire never established a lasting foothold in Scotland. For the real reason, see either Part 1 or here.

Custom

Some of the Institutional Writers mention “immemorial custom” as a source of law. Some rules of law are certainly “anterior to any statute and not comprehended in any, as being more solemn and sure than these are” (Stair, Institutions I, 1, 16). There is no statute, for example, that says murder is a crime, or that a person’s spouse and children are protected from being disinherited, yet both of these are rules of Scots law. The problem with using custom as a source of law is that it is quite difficult to find any customs that are old enough to be authoritative, but have not yet been brought up in court (since then they would either be part of the common law, meaning there’s a better source than custom; or they would be debunked). And until it’s tested in court, you don’t know whether the courts will respect it or not. The last time a custom was decided in court to be the law was in the 19th century. However, custom as a source of law is useful in Orkney and Shetland, where some practices deriving from Viking law (mostly to do with landownership without a title deed) have survived to the present day.

Juridical typology

That completes the list of sources of Scots law. So now for the promised brief word on typology. Broadly speaking, the world’s legal systems are divided into two groups: “common law” and “civil law” jurisdictions. Essentially, the common law jurisdictions are England & Wales and those countries whose law is descended from English law due to having been British colonies: e.g. 49 out of 50 US states, nine of out ten Canadian provinces, Australia, New Zealand, and both halves of Ireland. Their defining features are reliance on judicial precedent (see above) and a division between law and equity inherited from English law (mentioned in Part 1).

The civil law jurisdictions are generally continental European countries with Roman-inspired civil codes, their former colonies, and other countries that have copied the idea. And then there are the mixed legal systems, which have elements of both types. Usually this is due to influence from each of the other camps at different times, typically due to having been a colony both of Britain and of France or the Netherlands at different points in history (e.g. South Africa, Sri Lanka) or being the only civil law jurisdiction in a federation full of common law ones, so that the local law is civil but the federal law is common (e.g. Lousiana, Québec).

Scotland is always classed as a mixed legal system, but I have a pet theory that it isn’t really. Instead, the apparent mixture comes from the fact that the labels of “civil law” and “common law” jurisdictions have two different meanings. The first is about where the basic concepts originally come from. As I mentioned in Part 1, the English usually came up with their own solutions to problems when everyone else was borrowing them from the Romans. The concepts thereby created have been inherited by English law’s descendants and are lacking in other legal systems. So in one sense the question “is this a common law jurisdiction?” means “is it descended from English law?”, to which the answer for Scots law is undoubtedly no. But the civil law vs common law dichotomy is also about whether the underlying principles come from a civil code or from judicial precedents. In this sense Scotland is most definitely a common law jurisdiction. This kind of “mixture” comes not from alternate English and continental influence – Scotland used judicial precedent long before the Union – nor from split levels of local and federal law as in Louisiana and Québec – as previously mentioned, there is no “UK law”. Rather, it comes simply from people not being clear what they mean when they use the two terms. Or at least, so say I. If anyone knows of a paper either supporting or contradicting this idea, I’d love to know.

Next time: why there are not two but three possible verdicts in a Scottish criminal trial, and the safeguard against false convictions that no other country has.